Shrinking classroom age variance raises student achievement : evidence from developing countries

Large classroom variance of student age is prevalent in developing countries, where achievement tends to be low. This paper investigates whether increased classroom age variance adversely affects mathematics and science achievement. Using exogenous variation in the variance of student age in ability-mixing schools, the author finds robust negative effects of classroom age variance on fourth graders'achievement in developing countries. A simulation demonstrates that re-grouping students by age in the sample can improve math and science test scores by roughly 0.1 standard deviations. According to past estimates for the United States, this effect size is similar to that of raising expenditures per student by 26 percent.